HIV infection may adversely affect clinical response to chloroquine therapy for uncomplicated malaria in children

Malaria and HIV infection are each responsible for significant morbidity and mortality in sub-Saharan Africa. An interaction between these infections could have far-reaching adverse public health consequences. In contrast to previous reports ^[1], a recent study ^[2] has shown that HIV infection increases the risk of developing clinical malaria and parasitemia. Despite increasing resistance, chloroquine remains the first-line therapy for malaria in most African countries. Impaired immunity resulting from HIV infection could lead to a decreased response to chloroquine, rendering this relatively safe and inexpensive antimalarial agent ineffective. In this study, we assessed the clinical response to chloroquine therapy among HIV-infected and uninfected individuals. The study protocol was reviewed and approved by the Ugandan National AIDS Research Subcommittee.

The subjects were participants of two recently completed studies conducted at Old Mulago Hill dispensary in Kampala, Uganda. Full details of the studies have been described previously ^[3,4]. The first study ^[3] completed follow-up of 258 out of 294 (87.8%) patients and was designed and powered to characterize predictors of chloroquine treatment failure after standard dosing. The study did not originally include HIV serostatus as a predictor of treatment failure. The second study ^[4] was designed and powered to compare the efficacy of standard chloroquine (follow-up 94 out of 108 patients, 87.0%) and sulfadoxine-pyrimethamine (follow-up 93 out of 106 patients, 87.7%) for the treatment of uncomplicated malaria. In each study, consecutive patients aged 6 months and older with uncomplicated falciparum malaria at a density of 2000/μl or greater were recruited, treated, and followed for 14 days or until clinical failure. The clinical response to therapy was assessed using the standard World Health Organization criteria.

The present study is a secondary analysis entailing retrospective HIV-1 antibody testing of stored blood specimens from patients who received standard chloroquine therapy (25 mg/kg). Specimens were available for HIV antibody testing for 349 of the 352 (99.1%) subjects who received standard chloroquine therapy and completed follow-up. Specimens were tested using the Abbot Determine Screening test (Abbott Laboratories, Abbott Park, IL, USA); positive results were confirmed by the Trinity Uni-Gold test (Trinity Biotech plc, Bray, Ireland).

Of 192 specimens available for children under 5 years of age (age range 6-48 months, median 12 months, interquartile range 7-36 months), six (3.1%) were HIV positive. Of 157 specimens from subjects 5 years and older (age range 5-60 years, median 10 years, interquartile range 7-19 years), 23 (14.7%) were HIV positive. Early or late clinical failure to chloroquine therapy occurred in 126 (65.6%) children under 5 years of age compared with 45 (28.6%) subjects aged 5 years and older (chi-square test, P < 0.001). Six out of six (100%) HIV-positive children under 5 years of age failed treatment compared with 120 out of 186 (64.5%) HIV-negative children (Fisher's exact test, P = 0.096). No association between HIV serostatus and clinical treatment failure was observed among subjects 5 years and older (Fisher's exact test, P = 0.224).

These data suggest that treatment of falciparum malaria with chloroquine may be less effective among HIV-positive children under 5 years of age, compared with HIV-negative children, but should be confirmed in larger studies.

Moses R. Kamyaa

Catherine N. Kigonyaa

Willi McFarlandb